Gitiles
Code Review Sign In
gerrit-public.fairphone.software / platform / external / python / pycparser / f86dea169a1568afd44ed3436e93b1718bac7ad1 / . / README.html
blob: f0c9fad698656f4807090466a48fd810cf15fc85 [file] [log] [blame]
eli.bendersky347497d2010-12-10 15:17:28 +0200[diff] [blame]1<?xml version="1.0" encoding="utf-8" ?>
2<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
3<html xmlns="http://www.w3.org/1999/xhtml" xml:lang="en" lang="en">
4<head>
5<meta http-equiv="Content-Type" content="text/html; charset=utf-8" />
eli.benderskye91008a2011-10-16 06:21:59 +0200[diff] [blame]6<meta name="generator" content="Docutils 0.9: http://docutils.sourceforge.net/" />
7<title>pycparser v2.05</title>
eli.bendersky347497d2010-12-10 15:17:28 +0200[diff] [blame]8<meta name="author" content="Eli Bendersky" />
9<style type="text/css">
10
11/*
12:Author: David Goodger (goodger@python.org)
eli.benderskye91008a2011-10-16 06:21:59 +0200[diff] [blame]13:Id: $Id: html4css1.css 7056 2011-06-17 10:50:48Z milde $
eli.bendersky347497d2010-12-10 15:17:28 +0200[diff] [blame]14:Copyright: This stylesheet has been placed in the public domain.
15
16Default cascading style sheet for the HTML output of Docutils.
17
18See http://docutils.sf.net/docs/howto/html-stylesheets.html for how to
19customize this style sheet.
20*/
21
22/* used to remove borders from tables and images */
23.borderless, table.borderless td, table.borderless th {
24 border: 0 }
25
26table.borderless td, table.borderless th {
27 /* Override padding for "table.docutils td" with "! important".
28 The right padding separates the table cells. */
29 padding: 0 0.5em 0 0 ! important }
30
31.first {
32 /* Override more specific margin styles with "! important". */
33 margin-top: 0 ! important }
34
35.last, .with-subtitle {
36 margin-bottom: 0 ! important }
37
38.hidden {
39 display: none }
40
41a.toc-backref {
42 text-decoration: none ;
43 color: black }
44
45blockquote.epigraph {
46 margin: 2em 5em ; }
47
48dl.docutils dd {
49 margin-bottom: 0.5em }
50
eli.benderskye91008a2011-10-16 06:21:59 +0200[diff] [blame]51object[type="image/svg+xml"], object[type="application/x-shockwave-flash"] {
52 overflow: hidden;
53}
54
eli.bendersky347497d2010-12-10 15:17:28 +0200[diff] [blame]55/* Uncomment (and remove this text!) to get bold-faced definition list terms
56dl.docutils dt {
57 font-weight: bold }
58*/
59
60div.abstract {
61 margin: 2em 5em }
62
63div.abstract p.topic-title {
64 font-weight: bold ;
65 text-align: center }
66
67div.admonition, div.attention, div.caution, div.danger, div.error,
68div.hint, div.important, div.note, div.tip, div.warning {
69 margin: 2em ;
70 border: medium outset ;
71 padding: 1em }
72
73div.admonition p.admonition-title, div.hint p.admonition-title,
74div.important p.admonition-title, div.note p.admonition-title,
75div.tip p.admonition-title {
76 font-weight: bold ;
77 font-family: sans-serif }
78
79div.attention p.admonition-title, div.caution p.admonition-title,
80div.danger p.admonition-title, div.error p.admonition-title,
81div.warning p.admonition-title {
82 color: red ;
83 font-weight: bold ;
84 font-family: sans-serif }
85
86/* Uncomment (and remove this text!) to get reduced vertical space in
87 compound paragraphs.
88div.compound .compound-first, div.compound .compound-middle {
89 margin-bottom: 0.5em }
90
91div.compound .compound-last, div.compound .compound-middle {
92 margin-top: 0.5em }
93*/
94
95div.dedication {
96 margin: 2em 5em ;
97 text-align: center ;
98 font-style: italic }
99
100div.dedication p.topic-title {
101 font-weight: bold ;
102 font-style: normal }
103
104div.figure {
105 margin-left: 2em ;
106 margin-right: 2em }
107
108div.footer, div.header {
109 clear: both;
110 font-size: smaller }
111
112div.line-block {
113 display: block ;
114 margin-top: 1em ;
115 margin-bottom: 1em }
116
117div.line-block div.line-block {
118 margin-top: 0 ;
119 margin-bottom: 0 ;
120 margin-left: 1.5em }
121
122div.sidebar {
123 margin: 0 0 0.5em 1em ;
124 border: medium outset ;
125 padding: 1em ;
126 background-color: #ffffee ;
127 width: 40% ;
128 float: right ;
129 clear: right }
130
131div.sidebar p.rubric {
132 font-family: sans-serif ;
133 font-size: medium }
134
135div.system-messages {
136 margin: 5em }
137
138div.system-messages h1 {
139 color: red }
140
141div.system-message {
142 border: medium outset ;
143 padding: 1em }
144
145div.system-message p.system-message-title {
146 color: red ;
147 font-weight: bold }
148
149div.topic {
150 margin: 2em }
151
152h1.section-subtitle, h2.section-subtitle, h3.section-subtitle,
153h4.section-subtitle, h5.section-subtitle, h6.section-subtitle {
154 margin-top: 0.4em }
155
156h1.title {
157 text-align: center }
158
159h2.subtitle {
160 text-align: center }
161
162hr.docutils {
163 width: 75% }
164
eli.benderskye91008a2011-10-16 06:21:59 +0200[diff] [blame]165img.align-left, .figure.align-left, object.align-left {
eli.bendersky347497d2010-12-10 15:17:28 +0200[diff] [blame]166 clear: left ;
167 float: left ;
168 margin-right: 1em }
169
eli.benderskye91008a2011-10-16 06:21:59 +0200[diff] [blame]170img.align-right, .figure.align-right, object.align-right {
eli.bendersky347497d2010-12-10 15:17:28 +0200[diff] [blame]171 clear: right ;
172 float: right ;
173 margin-left: 1em }
174
eli.benderskye91008a2011-10-16 06:21:59 +0200[diff] [blame]175img.align-center, .figure.align-center, object.align-center {
176 display: block;
177 margin-left: auto;
178 margin-right: auto;
179}
180
eli.bendersky347497d2010-12-10 15:17:28 +0200[diff] [blame]181.align-left {
182 text-align: left }
183
184.align-center {
185 clear: both ;
186 text-align: center }
187
188.align-right {
189 text-align: right }
190
191/* reset inner alignment in figures */
192div.align-right {
eli.benderskye91008a2011-10-16 06:21:59 +0200[diff] [blame]193 text-align: inherit }
eli.bendersky347497d2010-12-10 15:17:28 +0200[diff] [blame]194
195/* div.align-center * { */
196/* text-align: left } */
197
198ol.simple, ul.simple {
199 margin-bottom: 1em }
200
201ol.arabic {
202 list-style: decimal }
203
204ol.loweralpha {
205 list-style: lower-alpha }
206
207ol.upperalpha {
208 list-style: upper-alpha }
209
210ol.lowerroman {
211 list-style: lower-roman }
212
213ol.upperroman {
214 list-style: upper-roman }
215
216p.attribution {
217 text-align: right ;
218 margin-left: 50% }
219
220p.caption {
221 font-style: italic }
222
223p.credits {
224 font-style: italic ;
225 font-size: smaller }
226
227p.label {
228 white-space: nowrap }
229
230p.rubric {
231 font-weight: bold ;
232 font-size: larger ;
233 color: maroon ;
234 text-align: center }
235
236p.sidebar-title {
237 font-family: sans-serif ;
238 font-weight: bold ;
239 font-size: larger }
240
241p.sidebar-subtitle {
242 font-family: sans-serif ;
243 font-weight: bold }
244
245p.topic-title {
246 font-weight: bold }
247
248pre.address {
249 margin-bottom: 0 ;
250 margin-top: 0 ;
251 font: inherit }
252
eli.benderskye91008a2011-10-16 06:21:59 +0200[diff] [blame]253pre.literal-block, pre.doctest-block, pre.math {
eli.bendersky347497d2010-12-10 15:17:28 +0200[diff] [blame]254 margin-left: 2em ;
255 margin-right: 2em }
256
257span.classifier {
258 font-family: sans-serif ;
259 font-style: oblique }
260
261span.classifier-delimiter {
262 font-family: sans-serif ;
263 font-weight: bold }
264
265span.interpreted {
266 font-family: sans-serif }
267
268span.option {
269 white-space: nowrap }
270
271span.pre {
272 white-space: pre }
273
274span.problematic {
275 color: red }
276
277span.section-subtitle {
278 /* font-size relative to parent (h1..h6 element) */
279 font-size: 80% }
280
281table.citation {
282 border-left: solid 1px gray;
283 margin-left: 1px }
284
285table.docinfo {
286 margin: 2em 4em }
287
288table.docutils {
289 margin-top: 0.5em ;
290 margin-bottom: 0.5em }
291
292table.footnote {
293 border-left: solid 1px black;
294 margin-left: 1px }
295
296table.docutils td, table.docutils th,
297table.docinfo td, table.docinfo th {
298 padding-left: 0.5em ;
299 padding-right: 0.5em ;
300 vertical-align: top }
301
302table.docutils th.field-name, table.docinfo th.docinfo-name {
303 font-weight: bold ;
304 text-align: left ;
305 white-space: nowrap ;
306 padding-left: 0 }
307
308h1 tt.docutils, h2 tt.docutils, h3 tt.docutils,
309h4 tt.docutils, h5 tt.docutils, h6 tt.docutils {
310 font-size: 100% }
311
312ul.auto-toc {
313 list-style-type: none }
314
315</style>
316</head>
317<body>
eli.benderskye91008a2011-10-16 06:21:59 +0200[diff] [blame]318<div class="document" id="pycparser-v2-05">
319<h1 class="title">pycparser v2.05</h1>
eli.bendersky347497d2010-12-10 15:17:28 +0200[diff] [blame]320<table class="docinfo" frame="void" rules="none">
321<col class="docinfo-name" />
322<col class="docinfo-content" />
323<tbody valign="top">
324<tr><th class="docinfo-name">Author:</th>
325<td><a class="first reference external" href="http://eli.thegreenplace.net">Eli Bendersky</a></td></tr>
326</tbody>
327</table>
328<div class="contents topic" id="contents">
329<p class="topic-title first">Contents</p>
330<ul class="auto-toc simple">
331<li><a class="reference internal" href="#introduction" id="id2">1&nbsp;&nbsp;&nbsp;Introduction</a><ul class="auto-toc">
332<li><a class="reference internal" href="#what-is-pycparser" id="id3">1.1&nbsp;&nbsp;&nbsp;What is pycparser?</a></li>
333<li><a class="reference internal" href="#what-is-it-good-for" id="id4">1.2&nbsp;&nbsp;&nbsp;What is it good for?</a></li>
334<li><a class="reference internal" href="#which-version-of-c-does-pycparser-support" id="id5">1.3&nbsp;&nbsp;&nbsp;Which version of C does pycparser support?</a></li>
335<li><a class="reference internal" href="#what-grammar-does-pycparser-follow" id="id6">1.4&nbsp;&nbsp;&nbsp;What grammar does pycparser follow?</a></li>
eli.bendersky40d218b2011-02-10 07:04:47 +0200[diff] [blame]336<li><a class="reference internal" href="#how-is-pycparser-licensed" id="id7">1.5&nbsp;&nbsp;&nbsp;How is pycparser licensed?</a></li>
337<li><a class="reference internal" href="#contact-details" id="id8">1.6&nbsp;&nbsp;&nbsp;Contact details</a></li>
eli.bendersky347497d2010-12-10 15:17:28 +0200[diff] [blame]338</ul>
339</li>
eli.bendersky40d218b2011-02-10 07:04:47 +0200[diff] [blame]340<li><a class="reference internal" href="#installing" id="id9">2&nbsp;&nbsp;&nbsp;Installing</a><ul class="auto-toc">
341<li><a class="reference internal" href="#prerequisites" id="id10">2.1&nbsp;&nbsp;&nbsp;Prerequisites</a></li>
342<li><a class="reference internal" href="#installation-process" id="id11">2.2&nbsp;&nbsp;&nbsp;Installation process</a></li>
343<li><a class="reference internal" href="#known-problems" id="id12">2.3&nbsp;&nbsp;&nbsp;Known problems</a></li>
eli.bendersky347497d2010-12-10 15:17:28 +0200[diff] [blame]344</ul>
345</li>
eli.bendersky40d218b2011-02-10 07:04:47 +0200[diff] [blame]346<li><a class="reference internal" href="#using" id="id13">3&nbsp;&nbsp;&nbsp;Using</a><ul class="auto-toc">
347<li><a class="reference internal" href="#interaction-with-the-c-preprocessor" id="id14">3.1&nbsp;&nbsp;&nbsp;Interaction with the C preprocessor</a></li>
348<li><a class="reference internal" href="#what-about-the-standard-c-library-headers" id="id15">3.2&nbsp;&nbsp;&nbsp;What about the standard C library headers?</a></li>
349<li><a class="reference internal" href="#basic-usage" id="id16">3.3&nbsp;&nbsp;&nbsp;Basic usage</a></li>
350<li><a class="reference internal" href="#advanced-usage" id="id17">3.4&nbsp;&nbsp;&nbsp;Advanced usage</a></li>
eli.bendersky347497d2010-12-10 15:17:28 +0200[diff] [blame]351</ul>
352</li>
eli.bendersky40d218b2011-02-10 07:04:47 +0200[diff] [blame]353<li><a class="reference internal" href="#modifying" id="id18">4&nbsp;&nbsp;&nbsp;Modifying</a></li>
354<li><a class="reference internal" href="#package-contents" id="id19">5&nbsp;&nbsp;&nbsp;Package contents</a></li>
355<li><a class="reference internal" href="#contributors" id="id20">6&nbsp;&nbsp;&nbsp;Contributors</a></li>
356<li><a class="reference internal" href="#changelog" id="id21">7&nbsp;&nbsp;&nbsp;Changelog</a></li>
eli.bendersky347497d2010-12-10 15:17:28 +0200[diff] [blame]357</ul>
358</div>
359<div class="section" id="introduction">
360<h1>1&nbsp;&nbsp;&nbsp;Introduction</h1>
361<div class="section" id="what-is-pycparser">
362<h2>1.1&nbsp;&nbsp;&nbsp;What is pycparser?</h2>
eli.benderskye91008a2011-10-16 06:21:59 +0200[diff] [blame]363<p><tt class="docutils literal">pycparser</tt> is a parser for the C language, written in pure Python. It is a module designed to be easily integrated into applications that need to parse C source code.</p>
eli.bendersky347497d2010-12-10 15:17:28 +0200[diff] [blame]364</div>
365<div class="section" id="what-is-it-good-for">
366<h2>1.2&nbsp;&nbsp;&nbsp;What is it good for?</h2>
eli.benderskye91008a2011-10-16 06:21:59 +0200[diff] [blame]367<p>Anything that needs C code to be parsed. The following are some uses for <tt class="docutils literal">pycparser</tt>, taken from real user reports:</p>
eli.bendersky347497d2010-12-10 15:17:28 +0200[diff] [blame]368<ul class="simple">
369<li>C code obfuscator</li>
370<li>Front-end for various specialized C compilers</li>
371<li>Static code checker</li>
372<li>Automatic unit-test discovery</li>
373<li>Adding specialized extensions to the C language</li>
374</ul>
eli.benderskye91008a2011-10-16 06:21:59 +0200[diff] [blame]375<p><tt class="docutils literal">pycparser</tt> is unique in the sense that it's written in pure Python - a very high level language that's easy to experiment with and tweak. To people familiar with Lex and Yacc, <tt class="docutils literal">pycparser</tt>'s code will be simple to understand.</p>
eli.bendersky347497d2010-12-10 15:17:28 +0200[diff] [blame]376</div>
377<div class="section" id="which-version-of-c-does-pycparser-support">
378<h2>1.3&nbsp;&nbsp;&nbsp;Which version of C does pycparser support?</h2>
eli.benderskye91008a2011-10-16 06:21:59 +0200[diff] [blame]379<p><tt class="docutils literal">pycparser</tt> aims to support the full C99 language (according to the standard ISO/IEC 9899). This is a new feature in the version 2.x series - earlier versions only supported C89. For more information on the change, read <a class="reference external" href="http://code.google.com/p/pycparser/wiki/C99support">this wiki page</a>.</p>
380<p><tt class="docutils literal">pycparser</tt> doesn't support any GCC extensions.</p>
eli.bendersky347497d2010-12-10 15:17:28 +0200[diff] [blame]381</div>
382<div class="section" id="what-grammar-does-pycparser-follow">
383<h2>1.4&nbsp;&nbsp;&nbsp;What grammar does pycparser follow?</h2>
eli.benderskye91008a2011-10-16 06:21:59 +0200[diff] [blame]384<p><tt class="docutils literal">pycparser</tt> very closely follows the C grammar provided in the end of the C99 standard document</p>
eli.bendersky347497d2010-12-10 15:17:28 +0200[diff] [blame]385</div>
eli.bendersky347497d2010-12-10 15:17:28 +0200[diff] [blame]386<div class="section" id="how-is-pycparser-licensed">
eli.bendersky40d218b2011-02-10 07:04:47 +0200[diff] [blame]387<h2>1.5&nbsp;&nbsp;&nbsp;How is pycparser licensed?</h2>
eli.bendersky15cb2702011-04-29 09:01:33 +0300[diff] [blame]388<p><a class="reference external" href="http://www.opensource.org/licenses/bsd-license.php">New BSD License</a></p>
eli.bendersky347497d2010-12-10 15:17:28 +0200[diff] [blame]389</div>
390<div class="section" id="contact-details">
eli.bendersky40d218b2011-02-10 07:04:47 +0200[diff] [blame]391<h2>1.6&nbsp;&nbsp;&nbsp;Contact details</h2>
eli.benderskye91008a2011-10-16 06:21:59 +0200[diff] [blame]392<p>Drop me an email to <a class="reference external" href="mailto:eliben&#64;gmail.com">eliben&#64;gmail.com</a> for any questions regarding <tt class="docutils literal">pycparser</tt>. For reporting problems with <tt class="docutils literal">pycparser</tt> or submitting feature requests, the best way is to open an issue on the <a class="reference external" href="http://code.google.com/p/pycparser/">pycparser page at Google Code</a>.</p>
eli.bendersky347497d2010-12-10 15:17:28 +0200[diff] [blame]393</div>
394</div>
395<div class="section" id="installing">
396<h1>2&nbsp;&nbsp;&nbsp;Installing</h1>
397<div class="section" id="prerequisites">
398<h2>2.1&nbsp;&nbsp;&nbsp;Prerequisites</h2>
399<ul class="simple">
eli.benderskye91008a2011-10-16 06:21:59 +0200[diff] [blame]400<li><tt class="docutils literal">pycparser</tt> was tested on Python 2.6 and 3.2, on both Linux and Windows. It should work on any later version (in both the 2.x and 3.x lines) as well.</li>
401<li><tt class="docutils literal">pycparser</tt> uses the PLY module for the actual lexer and parser construction. Install PLY from <a class="reference external" href="http://www.dabeaz.com/ply/">its website</a>.</li>
eli.bendersky347497d2010-12-10 15:17:28 +0200[diff] [blame]402</ul>
403</div>
404<div class="section" id="installation-process">
405<h2>2.2&nbsp;&nbsp;&nbsp;Installation process</h2>
eli.benderskye91008a2011-10-16 06:21:59 +0200[diff] [blame]406<p>Installing <tt class="docutils literal">pycparser</tt> is very simple. Once you download it from its <a class="reference external" href="http://code.google.com/p/pycparser/">website</a> and unzip the package, you just have to execute the standard <tt class="docutils literal">python setup.py install</tt>. The setup script will then place the <tt class="docutils literal">pycparser</tt> module into <tt class="docutils literal"><span class="pre">site-packages</span></tt> in your Python's installation library.</p>
407<p>Alternatively, since <tt class="docutils literal">pycparser</tt> is listed in the <a class="reference external" href="http://pypi.python.org/pypi/pycparser">Python Package Index</a> (PyPI), you can install it using your favorite Python packaging/distribution tool, for example with:</p>
408<pre class="literal-block">
409&gt; pip install pycparser
410</pre>
411<p>It's recommended to run <tt class="docutils literal">_build_tables.py</tt> in the <tt class="docutils literal">pycparser</tt> code directory after installation to make sure the parsing tables of PLY are pre-generated. This can make your code run faster.</p>
eli.bendersky347497d2010-12-10 15:17:28 +0200[diff] [blame]412</div>
413<div class="section" id="known-problems">
414<h2>2.3&nbsp;&nbsp;&nbsp;Known problems</h2>
415<ul class="simple">
eli.benderskye91008a2011-10-16 06:21:59 +0200[diff] [blame]416<li>Some users who've installed a new version of <tt class="docutils literal">pycparser</tt> over an existing version ran into a problem using the newly installed library. This has to do with parse tables staying around as <tt class="docutils literal">.pyc</tt> files from the older version. If you see unexplained errors from <tt class="docutils literal">pycparser</tt> after an upgrade, remove it (by deleting the <tt class="docutils literal">pycparser</tt> directory in your Python's <tt class="docutils literal"><span class="pre">site-packages</span></tt>, or wherever you installed it) and install again.</li>
eli.bendersky347497d2010-12-10 15:17:28 +0200[diff] [blame]417</ul>
418</div>
419</div>
420<div class="section" id="using">
421<h1>3&nbsp;&nbsp;&nbsp;Using</h1>
422<div class="section" id="interaction-with-the-c-preprocessor">
423<h2>3.1&nbsp;&nbsp;&nbsp;Interaction with the C preprocessor</h2>
eli.benderskye91008a2011-10-16 06:21:59 +0200[diff] [blame]424<p>In order to be compilable, C code must be preprocessed by the C preprocessor - <tt class="docutils literal">cpp</tt>. <tt class="docutils literal">cpp</tt> handles preprocessing directives like <tt class="docutils literal">#include</tt> and <tt class="docutils literal">#define</tt>, removes comments, and does other minor tasks that prepare the C code for compilation.</p>
425<p>For all but the most trivial snippets of C code, <tt class="docutils literal">pycparser</tt>, like a C compiler, must receive preprocessed C code in order to function correctly. If you import the top-level <tt class="docutils literal">parse_file</tt> function from the <tt class="docutils literal">pycparser</tt> package, it will interact with <tt class="docutils literal">cpp</tt> for you, as long as it's in your PATH, or you provide a path to it.</p>
426<p>On the vast majority of Linux systems, <tt class="docutils literal">cpp</tt> is installed and is in the PATH. If you're on Windows and don't have <tt class="docutils literal">cpp</tt> somewhere, you can use the one provided in the <tt class="docutils literal">utils</tt> directory in <tt class="docutils literal">pycparser</tt>'s distribution. This <tt class="docutils literal">cpp</tt> executable was compiled from the <a class="reference external" href="http://www.cs.princeton.edu/software/lcc/">LCC distribution</a>, and is provided under LCC's license terms.</p>
eli.bendersky347497d2010-12-10 15:17:28 +0200[diff] [blame]427</div>
428<div class="section" id="what-about-the-standard-c-library-headers">
429<h2>3.2&nbsp;&nbsp;&nbsp;What about the standard C library headers?</h2>
eli.benderskye91008a2011-10-16 06:21:59 +0200[diff] [blame]430<p>C code almost always includes various header files from the standard C library, like <tt class="docutils literal">stdio.h</tt>. While, with some effort, <tt class="docutils literal">pycparser</tt> can be made to parse the standard headers from any C compiler, it's much simpler to use the provided &quot;fake&quot; standard includes in <tt class="docutils literal">utils/fake_libc_include</tt>. These are standard C header files that contain only the bare necessities to allow valid parsing of the files that use them. As a bonus, since they're minimal, it can significantly improve the performance of parsing C files.</p>
431<p>See the <tt class="docutils literal">using_cpp_libc.py</tt> example for more details.</p>
eli.bendersky347497d2010-12-10 15:17:28 +0200[diff] [blame]432</div>
433<div class="section" id="basic-usage">
434<h2>3.3&nbsp;&nbsp;&nbsp;Basic usage</h2>
eli.benderskye91008a2011-10-16 06:21:59 +0200[diff] [blame]435<p>Take a look at the <tt class="docutils literal">examples</tt> directory of the distribution for a few examples of using <tt class="docutils literal">pycparser</tt>. These should be enough to get you started.</p>
eli.bendersky347497d2010-12-10 15:17:28 +0200[diff] [blame]436</div>
437<div class="section" id="advanced-usage">
438<h2>3.4&nbsp;&nbsp;&nbsp;Advanced usage</h2>
eli.benderskye91008a2011-10-16 06:21:59 +0200[diff] [blame]439<p>The public interface of <tt class="docutils literal">pycparser</tt> is well documented with comments in <tt class="docutils literal">pycparser/c_parser.py</tt>. For a detailed overview of the various AST nodes created by the parser, see <tt class="docutils literal">pycparser/_c_ast.cfg</tt>.</p>
eli.bendersky40d218b2011-02-10 07:04:47 +0200[diff] [blame]440<p>There's also a <a class="reference external" href="http://code.google.com/p/pycparser/wiki/FAQ">FAQ available here</a>. In any case, you can always drop me an <a class="reference external" href="mailto:eliben&#64;gmail.com">email</a> for help.</p>
eli.bendersky347497d2010-12-10 15:17:28 +0200[diff] [blame]441</div>
442</div>
443<div class="section" id="modifying">
444<h1>4&nbsp;&nbsp;&nbsp;Modifying</h1>
eli.benderskye91008a2011-10-16 06:21:59 +0200[diff] [blame]445<p>There are a few points to keep in mind when modifying <tt class="docutils literal">pycparser</tt>:</p>
eli.bendersky347497d2010-12-10 15:17:28 +0200[diff] [blame]446<ul class="simple">
eli.benderskye91008a2011-10-16 06:21:59 +0200[diff] [blame]447<li>The code for <tt class="docutils literal">pycparser</tt>'s AST nodes is automatically generated from a configuration file - <tt class="docutils literal">_c_ast.cfg</tt>, by <tt class="docutils literal">_ast_gen.py</tt>. If you modify the AST configuration, make sure to re-generate the code.</li>
448<li>Make sure you understand the optimized mode of <tt class="docutils literal">pycparser</tt> - for that you must read the docstring in the constructor of the <tt class="docutils literal">CParser</tt> class. For development you should create the parser without optimizations, so that it will regenerate the Yacc and Lex tables when you change the grammar.</li>
eli.bendersky347497d2010-12-10 15:17:28 +0200[diff] [blame]449</ul>
450</div>
451<div class="section" id="package-contents">
452<h1>5&nbsp;&nbsp;&nbsp;Package contents</h1>
eli.benderskye91008a2011-10-16 06:21:59 +0200[diff] [blame]453<p>Once you unzip the <tt class="docutils literal">pycparser</tt> package, you'll see the following files and directories:</p>
eli.bendersky347497d2010-12-10 15:17:28 +0200[diff] [blame]454<dl class="docutils">
455<dt>README.txt/html:</dt>
456<dd>This README file.</dd>
457<dt>setup.py:</dt>
458<dd>Installation script</dd>
459<dt>examples/:</dt>
eli.benderskye91008a2011-10-16 06:21:59 +0200[diff] [blame]460<dd>A directory with some examples of using <tt class="docutils literal">pycparser</tt></dd>
eli.bendersky347497d2010-12-10 15:17:28 +0200[diff] [blame]461<dt>pycparser/:</dt>
eli.benderskye91008a2011-10-16 06:21:59 +0200[diff] [blame]462<dd>The <tt class="docutils literal">pycparser</tt> module source code.</dd>
eli.bendersky347497d2010-12-10 15:17:28 +0200[diff] [blame]463<dt>tests/:</dt>
464<dd>Unit tests.</dd>
465<dt>utils/cpp.exe:</dt>
466<dd>A Windows executable of the C pre-processor suitable for working with pycparser</dd>
467<dt>utils/fake_libc_include:</dt>
468<dd>Minimal standard C library include files that should allow to parse any C code.</dd>
469<dt>utils/internal/:</dt>
470<dd>Internal utilities for my own use. You probably don't need them.</dd>
471</dl>
472</div>
473<div class="section" id="contributors">
474<h1>6&nbsp;&nbsp;&nbsp;Contributors</h1>
eli.benderskye91008a2011-10-16 06:21:59 +0200[diff] [blame]475<p>Some people have contributed to <tt class="docutils literal">pycparser</tt> by opening issues on bugs they've found and/or submitting patches. The list of contributors is at <a class="reference external" href="http://code.google.com/p/pycparser/wiki/Contributors">this pycparser Wiki page</a>.</p>
eli.bendersky347497d2010-12-10 15:17:28 +0200[diff] [blame]476</div>
477<div class="section" id="changelog">
478<h1>7&nbsp;&nbsp;&nbsp;Changelog</h1>
479<ul class="simple">
eli.benderskye91008a2011-10-16 06:21:59 +0200[diff] [blame]480<li>Version 2.05 (16.10.2011)<ul>
481<li>Added support for the C99 <tt class="docutils literal">_Bool</tt> type and <tt class="docutils literal">stdbool.h</tt> header file</li>
482<li>Expanded <tt class="docutils literal">examples/explore_ast.py</tt> with more details on working with the
483AST</li>
484<li>Relaxed the rules on parsing unnamed struct members (helps parse <tt class="docutils literal">windows.h</tt>)</li>
485<li>Bug fixes:<ul>
486<li>Fixed spacing issue for some type declarations</li>
487<li>Issue 47: display empty statements (lone ';') correctly after parsing</li>
488</ul>
489</li>
490</ul>
491</li>
eli.bendersky851dbfb2011-05-21 09:30:05 +0300[diff] [blame]492<li>Version 2.04 (21.05.2011)<ul>
493<li>License changed from LGPL to BSD</li>
494<li>Bug fixes:<ul>
495<li>Issue 31: constraining the scope of typedef definitions</li>
496<li>Issues 33, 35: fixes for the c-to-c.py example</li>
497</ul>
498</li>
499<li>Added C99 integer types to fake headers</li>
500<li>Added unit tests for the c-to-c.py example</li>
501</ul>
502</li>
eli.benderskyb56e8f82011-03-06 07:56:55 +0200[diff] [blame]503<li>Version 2.03 (06.03.2011)<ul>
504<li>Bug fixes:<ul>
505<li>Issue 17: empty file-level declarations</li>
506<li>Issue 18: empty statements and declarations in functions</li>
507<li>Issue 19: anonymous structs &amp; union fields</li>
508<li>Issue 23: fix coordinates of Cast nodes</li>
509</ul>
510</li>
eli.benderskye91008a2011-10-16 06:21:59 +0200[diff] [blame]511<li>New example added (<tt class="docutils literal"><span class="pre">examples/c-to-c.py</span></tt>) for translating ASTs generated by <tt class="docutils literal">pycparser</tt> back into C code.</li>
512<li><tt class="docutils literal">pycparser</tt> is now on PyPI (Python Package Index)</li>
513<li>Created <a class="reference external" href="http://code.google.com/p/pycparser/wiki/FAQ">FAQ</a> on the <tt class="docutils literal">pycparser</tt> project page</li>
514<li>Removed support for Python 2.5. <tt class="docutils literal">pycparser</tt> supports Python 2 from 2.6 and on, and Python 3.</li>
eli.benderskyb56e8f82011-03-06 07:56:55 +0200[diff] [blame]515</ul>
516</li>
eli.bendersky347497d2010-12-10 15:17:28 +0200[diff] [blame]517<li>Version 2.02 (10.12.2010)<ul>
eli.benderskye91008a2011-10-16 06:21:59 +0200[diff] [blame]518<li>The name of a <tt class="docutils literal">NamedInitializer</tt> node was turned into a sequence of nodes
eli.bendersky347497d2010-12-10 15:17:28 +0200[diff] [blame]519instead of an attribute, to make it discoverable by the AST node visitor.</li>
520<li>Documentation updates</li>
521</ul>
522</li>
523<li>Version 2.01 (04.12.2010)<ul>
524<li>Removed dependency on YAML. Parsing of the AST node configuration file is done with a simple parser.</li>
525<li>Fixed issue 12: installation problems</li>
526</ul>
527</li>
528<li>Version 2.00 (31.10.2010)<ul>
529<li>Support for C99 (read <a class="reference external" href="http://code.google.com/p/pycparser/wiki/C99support">this wiki page</a> for more information).</li>
530</ul>
531</li>
532<li>Version 1.08 (09.10.2010)<ul>
533<li>Bug fixes:<ul>
eli.benderskye91008a2011-10-16 06:21:59 +0200[diff] [blame]534<li>Correct handling of <tt class="docutils literal">do{} ... while</tt> statements in some cases</li>
eli.bendersky347497d2010-12-10 15:17:28 +0200[diff] [blame]535<li>Issues 6 &amp; 7: Concatenation of string literals</li>
536<li>Issue 9: Support for unnamed bitfields in structs</li>
537</ul>
538</li>
539</ul>
540</li>
541<li>Version 1.07 (18.05.2010)<ul>
eli.benderskye91008a2011-10-16 06:21:59 +0200[diff] [blame]542<li>Python 3.1 compatibility: <tt class="docutils literal">pycparser</tt> was modified to run on Python 3.1 as well as 2.6</li>
eli.bendersky347497d2010-12-10 15:17:28 +0200[diff] [blame]543</ul>
544</li>
545<li>Version 1.06 (10.04.2010)<ul>
546<li>Bug fixes:<ul>
547<li>coord not propagated to FuncCall nodes</li>
548<li>lexing of the ^= token (XOREQUALS)</li>
549<li>parsing failed on some abstract declarator rules</li>
550</ul>
551</li>
eli.benderskye91008a2011-10-16 06:21:59 +0200[diff] [blame]552<li>Linux compatibility: fixed end-of-line and <tt class="docutils literal">cpp</tt> path issues to allow all tests and examples run on Linux</li>
eli.bendersky347497d2010-12-10 15:17:28 +0200[diff] [blame]553</ul>
554</li>
555<li>Version 1.05 (16.10.2009)<ul>
eli.benderskye91008a2011-10-16 06:21:59 +0200[diff] [blame]556<li>Fixed the <tt class="docutils literal">parse_file</tt> auxiliary function to handle multiple arguments to <tt class="docutils literal">cpp</tt> correctly</li>
eli.bendersky347497d2010-12-10 15:17:28 +0200[diff] [blame]557</ul>
558</li>
559<li>Version 1.04 (22.05.2009)<ul>
eli.benderskye91008a2011-10-16 06:21:59 +0200[diff] [blame]560<li>Added the <tt class="docutils literal">fake_libc_include</tt> directory to allow parsing of C code that uses standard C library include files without dependency on a real C library.</li>
eli.bendersky347497d2010-12-10 15:17:28 +0200[diff] [blame]561<li>Tested with Python 2.6 and PLY 3.2</li>
562</ul>
563</li>
564<li>Version 1.03 (31.01.2009)<ul>
565<li>Accept enumeration lists with a comma after the last item (C99 feature).</li>
566</ul>
567</li>
568<li>Version 1.02 (16.01.2009)<ul>
eli.benderskye91008a2011-10-16 06:21:59 +0200[diff] [blame]569<li>Fixed problem of parsing struct/enum/union names that were named similarly to previously defined <tt class="docutils literal">typedef</tt> types.</li>
eli.bendersky347497d2010-12-10 15:17:28 +0200[diff] [blame]570</ul>
571</li>
572<li>Version 1.01 (09.01.2009)<ul>
573<li>Fixed subprocess invocation in the helper function parse_file - now it's more portable</li>
574</ul>
575</li>
576<li>Version 1.0 (15.11.2008)<ul>
577<li>Initial release</li>
578<li>Support for ANSI C89</li>
579</ul>
580</li>
581</ul>
582</div>
583</div>
584</body>
585</html>